README.md

---
tags:
  - security
  - huntr
  - model-file
  - tensorrt
  - detectionlayer
  - backdoor
---

# TensorRT DetectionLayer silent suppression proof of concept

This repository contains a bounded research PoC for TensorRT engine files that
embed a `DetectionLayer_TRT` payload with semantically invalid serialized
parameters.

The security question is whether a normal engine load + inference path will
silently suppress detections compared with a benign control engine.

## Files

- `control.engine`
- `neg_keepTopK.engine`
- `verify_remote_poc.py`
- `requirements.txt`

## What the files do

Both files are valid TensorRT engine files that load via:

```python
runtime.deserialize_cuda_engine(...)
```

Both also execute via:

- `engine.create_execution_context()`
- `execute_async_v3(0)`

Control behavior:

```json
{
  "positive_cls": [0.5],
  "mixed_bbox": [0.11951626092195511]
}
```

Malicious behavior:

```json
{
  "positive_cls": [0.0],
  "mixed_bbox": [0.0]
}
```

## Verify the public HF artifacts

After unauthenticated download, run:

```bash
python verify_remote_poc.py
```

Expected result:

- both engines load successfully
- control and malicious execution both succeed
- the malicious engine suppresses outputs that are present in the control engine

## Safety note

This is a bounded research PoC:

- no code execution claim
- no external callbacks
- only deterministic execute-time output suppression after a trusted engine load